Epstein-Barr virus (EBV) is associated with B-cell malignancies in AIDS patients, and new treatments are urgently needed to treat AIDS patients with EBV-induced lymphomas. In addition, better animal models, with intact human immune systems, must be developed to define the roles of specific viral proteins in EBV-induced lymphomas, and to test new anti-tumor therapies. In Aim 1 of this proposal, we will use a newly developed humanized mouse model to clarify the role(s) of two EBV latent proteins (LMP1 and LMP2A) for the establishment of lymphomas in the context of an intact viral genome, and in the presence or absence of a functional immune system. Since our unexpected and intriguing new preliminary results indicate that both the LMP1 and LMP2A EBV mutants can induce B cell lymphomas in humanized mice when T cell function is inhibited, we hypothesize that the transforming functions of LMP1 and LMP2A are partially redundant in this model, and that the quality of the T cell response affects the roles of LMP1 and LMP2A. In Aim 2, we will examine the transforming phenotype of a double LMP1/LMP2A mutant in the humanized mouse model, and use this model (as well as in vitro studies) to explore the effects of LMP1 versus LMP2A on viral reactivation. We hypothesize that the double LMP1/LMP2A mutant virus will be completely defective for lymphoma formation in humanized mice, and that LMP1 and LMP2A have opposing effects on lytic viral reactivation. Finally, in Aim 3, we will explore the potential use f the FDA-approved arthritis drug, leflunomide, as a treatment for EBV+ AIDS-related lymphomas. Our exciting preliminary results show that leflunomide not only induces lytic EBV reactivation through an ATM-dependent mechanism, but that it also kills many different types of EBV-transformed B cells in vitro (including EBV+ Burkitt lymphomas and lymphoblastoid cell lines) at clinically non-toxic doses. We hypothesize that the humanized mouse model will reveal that the roles of LMP1/LMP2A in EBV-induced lymphomas are different in the presence versus absence of functional T cells, and that leflunomide can be used to treat EBV-positive AIDS-related lymphomas both through its ability to induce ATM/p53-dependent lytic EBV reactivation, and to inhibit essential EBV-activated cellular pathways (including c-myc, NF-Kappa B, and STAT3) that are required for the viability of EBV-transformed B cells.